Generating Image Compositions

1. A computer-implemented method comprising: receiving a plurality of photos; determining a composition type from the plurality of photos, wherein the composition type is one of a plurality of different composition types, and wherein determining the composition type includes: determining that the plurality of photos meet one or more particular criteria for the composition type, wherein the one or more particular criteria relate to one or more comparisons between the plurality of photos; determining image content in the plurality of photos, the image content including foreground content and background content; and determining that at least a portion of the image content corresponds among the plurality of photos in a scene present in the plurality of the photos; generating a composition from the plurality of photos based on the composition type, wherein the generating includes modifying pixels of a base photo of the plurality of photos by adding respective portions of foreground content from different photos of the plurality of photos; and causing output of the composition by a display device.

2. The computer-implemented method of claim 1 , wherein determining that at least the portion of the image content corresponds among the plurality of photos includes comparing metadata of the plurality of photos.

3. The computer-implemented method of claim 2 , wherein comparing metadata includes comparing one or more of tags, timestamps, and geo-locations of the plurality of photos.

4. The computer-implemented method of claim 1 , wherein determining that the plurality of photos meet the one or more particular criteria includes at least one of: determining that the plurality of photos were captured in sequence; and determining that a position of corresponding foreground content in the different photos changes relative to corresponding background content.

5. The computer-implemented method of claim 4 , wherein the position of the corresponding foreground content is in a centrally-located portion of each of the plurality of photos.

6. The computer-implemented method of claim 1 , wherein the composition is of an action composition type, and wherein an active object is in at least two different positions in the composition.

7. A system comprising: one or more hardware processors; and logic encoded in one or more tangible media, that when executed by the one or more hardware processors causes the one or more hardware processors to perform operations comprising: receiving a plurality of photos; determining that the plurality of photos meet criteria for at least one composition type, wherein the criteria relate to one or more comparisons between the plurality of photos; determining image content in the plurality of photos, the image content including foreground content and background content; determining that at least a portion of the image content corresponds among the plurality of photos in a scene present in the plurality of the photos; generating a composition of the at least one composition type from the plurality of photos, wherein the generating includes modifying pixels of a base photo of the plurality of photos by adding respective portions of foreground content from different photos of the plurality of photos; and providing the composition to a user.

8. The system of claim 7 , wherein determining that at least the portion of the image content corresponds among the plurality of photos includes comparing metadata of the plurality of photos.

9. The system of claim 8 , wherein comparing metadata includes comparing one or more of tags, timestamps, and geo-locations of the plurality of photos.

10. The system of claim 7 , wherein determining that the plurality of photos meet the criteria includes at least one of: determining that the plurality of photos were captured in sequence; and determining that a position of corresponding foreground content in the different photos changes relative to corresponding background content.

11. The system of claim 10 , wherein the position of the corresponding foreground content is in a centrally-located portion of each of the plurality of photos.

12. The system of claim 7 , wherein the composition is of an action composition type, and wherein an active object is in at least two different positions in the composition.

13. The system of claim 7 , wherein the composition is of a clutter-removed composition type, and wherein the plurality of photos each include a foreground object that is in a different position in each of the plurality of photos, and wherein the composition excludes the foreground object.

14. A computer-implemented method comprising: determining a composition type from a plurality of photos, wherein determining the composition type includes: determining, based on comparison of the plurality of photos, that the plurality of photos meet one or more particular criteria for the composition type; determining image content in the plurality of photos, the image content including foreground content and background content; and determining that the image content in the plurality of the photos is similar; generating a composition from the plurality of photos based on the composition type, wherein the generating comprises modifying pixels of a base photo of the plurality of photos by adding a respective portion of image content from one or more other photos of the plurality of photos; and providing the composition to a user.

15. The computer-implemented method of claim 14 , wherein generating the composition further comprises: segmenting each of the plurality of photos; and based on the segmenting, determining the respective portion of the image content in the one or more other photos.

16. The computer-implemented method of claim 14 , wherein the adding comprises applying a blending algorithm to the respective portion of the image content of the one or more other photos.

17. The computer-implemented method of claim 14 , wherein determining that the image content in the plurality of the photos is similar comprises recognizing a same object in a center region of each of the plurality of photos.

18. The computer-implemented method of claim 17 , wherein the respective portion of the image content is associated with the same object, and wherein adding the respective portion of the image content comprises replacing a particular foreground object in the base photo with the respective portion.

19. The computer-implemented method of claim 14 , wherein determining that the image content in the plurality of the photos is similar comprises: recognizing a foreground object in each of the plurality of photos, wherein the foreground object is in a different position in each of the plurality of photos; and recognizing a static background in each of the plurality of photos.

20. The computer-implemented method of claim 19 , wherein the respective portion of the one or more other photos includes the foreground object.